ipv6: Always allocate pcpu memory in a fib6_nh
Subsystems (3)
David Ahern
maintainerPATCH 1/1 ipv6: Always allocate pcpu memory in a fib6_nh
From: David Ahern <[email protected]>
A recent commit had an unintended side effect with reject routes:rt6i_pcpu is expected to always be initialized for all fib6_info except the null entry. The commit mentioned below skips it for reject routes and ends up leaking references to the loopback device. For example,
ip netns add foo
ip -netns foo li set lo up
ip -netns foo -6 ro add blackhole 2001:db8:1::1 ip netns exec foo ping6 2001:db8:1::1 ip netns del foo
ends up spewing:
unregister_netdevice: waiting for lo to become free. Usage count = 3
The fib_nh_common_init is not needed for reject routes (no ipv4 caching or encaps), so move the alloc_percpu_gfp after it and adjust the goto label.
--- net/ipv6/route.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-)
net/ipv6/route.c +7 -6
@@ -3400,7 +3400,7 @@ int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh, goto out; } } - goto set_dev; + goto pcpu_alloc; }if (cfg->fc_flags &RTF_GATEWAY ) {
@@ -3432,17 +3432,18 @@ int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh, !netif_carrier_ok(dev)) fib6_nh->fib_nh_flags |=RTNH_F_LINKDOWN ; + err =fib_nh_common_init (&fib6_nh->nh_common, cfg->fc_encap, + cfg->fc_encap_type, cfg, gfp_flags, extack); +if (err) + goto out; + +pcpu_alloc: fib6_nh->rt6i_pcpu = alloc_percpu_gfp(struct rt6_info *, gfp_flags);if (!fib6_nh->rt6i_pcpu) { err = -ENOMEM ; goto out; } - err =fib_nh_common_init (&fib6_nh->nh_common, cfg->fc_encap, - cfg->fc_encap_type, cfg, gfp_flags, extack); -if (err) - goto out; -set_dev: fib6_nh->fib_nh_dev = dev; fib6_nh->fib_nh_oif = dev->ifindex; err =0 ;
Signature · 1 line
2.11.0
From: David Ahern <[email protected]>
Date: Mon, 3 Jun 2019 18:37:03 -0700
Quoted 13 lines
From: David Ahern <[email protected]>
A recent commit had an unintended side effect with reject routes:
rt6i_pcpuis expected to always be initialized for allfib6_infoexcept the null entry. The commit mentioned below skips it for reject routes and ends up leaking references to the loopback device. For example,ip netns add foo
ip -netns foo li set lo up
ip -netns foo -6 ro add blackhole 2001:db8:1::1 ip netns exec foo ping6 2001:db8:1::1 ip netns del foo
Quoted 8 lines
ends up spewing:
unregister_netdevice: waiting for lo to become free. Usage count = 3The
fib_nh_common_initis not needed for reject routes (no ipv4 caching or encaps), so move thealloc_percpu_gfpafter it and adjust the goto label.Fixes f40b6ae2b612 ("ipv6: Move pcpu cached routes to fib6_nh")Signed-off-by David Ahern <[email protected]>
Applied.